Adaptive non-cartesian networks for vision

  • J. R. Serra
  • J. Brian Subirana
Session 10: Recognition & Reconstruction
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1311)


We address the problem of locating and extracting frame curves on interesting image areas. Reference frames, focus of attention, bounding contours of shapes, axis of inertia, centers of masses and other mid-level visual structures, can be used to guide other mid-level visual tasks or to lead subsequent high level processing like recognition, indexing or image retrieval. Frame curves, are useful to tackle non-rigid object recognition problems because these have fuzzy boundaries. Where is the boundary of a cloud, oak leave or a leopard? We present a perceptual organization approach based on dynamic programming and adaptive non-cartesian networks, a new kind of networks which are based on placing processor lines using a distribution function adapted to the image array. We present a novel computational framework to extract frame curves directly on the image and several experiments on real images.


Image Retrieval Object Boundary Active Contour Model Dynamic Programming Approach Salient Area 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. [Amini, Tehrani and Weymouth 1988]
    Amir A. Amini, Saeid Tehrani and Terry E. Weymouth. Using dynamic programming for minimizing the energy of active contours in the presence of hard constraints. In Proceedings of Second International Conference on Computer Vision, pages 95–99, Florida 1988.Google Scholar
  2. [Alter and Basri 1995]
    T.D. Alter and R. Basri. Extracting salient curves from images: an analysis of the saliency network. MIT AI TR 1550, 1995.Google Scholar
  3. [Havaldar et al. 1994]
    P. Havaldar, G. Medioni and F. Stein. Extraction of groups for recognition. In Proceedings of the European Conference on Computer Vision, pages 251–261, 1994Google Scholar
  4. [Serra and Subirana 1996]
    J. R. Serra and J. B. Subirana. Perceptual Grouping on Texture Images using Non-Cartesian Networks. In Proceedings of 13th International Conference on Pattern Recognition. Vol. II, pages 462–466. Vienna, 1996.Google Scholar
  5. [Subirana-Vilanova 1993]
    J. B. Subirana-Vilanova. Middle level vision and visual recognition of non-rigid objects. MIT PhD thesis. August 1993.Google Scholar
  6. [Sha'ashua and Ullman 1988]
    A. Sha'ashua and S. Ullman. Structural saliency: The detection of globally salient structures using a locally connected network. Proceedings ICCV 2, pages 321–327, 1988.Google Scholar
  7. [Vasilescu and Terzopoulos 1992]
    M. Vasilescu and D. Terzopoulos. Adaptive meshes and shells: irregular triangulation, discontinuities, and hierarchical subdivision. In Proceedings of IEEE International Conference on Computer Vision and Pattern Recognition, pages 829–831, Illinois 1992.Google Scholar
  8. [Young 1985]
    R. Young. The gaussian derivative theory of spatial vision: Analysis of cortical cell receptive field line-weighting profiles. Technical Report GMR-4920. General Motors Research, 1985.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • J. R. Serra
    • 1
    • 2
  • J. Brian Subirana
    • 2
  1. 1.Dipartimento di Sistemi e InformaticaUniversità degli Studi di FirenzeFirenzeItaly
  2. 2.Departamento de Informatica. Edificio CUniversidad Autónoma de BarcelonaBarcelonaSpain

Personalised recommendations